The International Journal of Business and Finance Research ♦ Volume 2 ♦ Number 1 ♦ 2008

MARKET VALUATION RESPONSES TO GOODWILL ANNOUNCEMENTS: AN EARLY DIRECT TEST OF FASB 142 Halil Kiymaz, Rollins College Roberto Marchesini, University of Houston-Clear Lake Robert F. Hodgin, University of Houston-Clear Lake

ABSTRACT

This study provides evidence from an early direct test of the Financial Accounting Standards Board (FASB) 142 policy statement regarding investor market reaction to corporate goodwill impairment announcements. Under new rules, the amortization of goodwill is replaced with a two-step procedure to determine if goodwill is impaired. We draw a sample of 188 firms announcing impairment tests during the period of 2001-2003 to investigate market reactions. The findings for overall sample indicate that firms with impairment test announcements experience statistically significant negative abnormal returns. The findings further show that the abnormal returns are negative for NYSE and AMEX+NASDAQ listed stocks. When we analyze the industry affiliation of firms and abnormal returns based on the primary SIC of firms, we find varying market reactions to goodwill announcements among industries. The findings are consistent with an information efficiency view of the market and one other related study of similar design, conducted before the effective date of FASB 142.

JEL: G14, M41

INTRODUCTION

ith the June 2001 release of the Financial Accounting Standards Board (FASB) Statement 142, the accounting profession changed its criteria for goodwill reporting in corporate Wconsolidations. The statement aims to recast how to account for goodwill, and in so doing, improve both the assessment and reporting of goodwill’s economic value in the merged enterprise. The statement’s language shifts goodwill accounting from a fixed amortization of no more than 40 years, to an asset whose life span and value to the reporting unit must be tested annually. The profession expects that the new reporting criteria will more clearly reveal the combined enterprise’s goodwill value to both analysts and investors.

The 1980’s merger and acquisition wave and the growing importance of goodwill as a percentage of post- merger assets re-focused the accounting profession’s attention on goodwill’s value and measurement. Beginning in 1970, Accounting Principles Board (APB) Opinion 16 had provided for two different post- merger goodwill reporting means: the pooling-of-interests method or the purchase method. By permitting alternative goodwill accounting methods, earnings reports for merged operations could differ based on accounting technique alone. The work of making equivalent financial statement comparisons between companies was left to analysts and investors.

Relief from this reporting dualism appeared in FASB Statement 141. Effective for all business combinations after June 30, 2001, that language mandated all post-merger goodwill reporting use the purchase method. The move to a single goodwill accounting method reduced reporting inconsistencies, but left untouched the rather artificial treatment of goodwill as an intangible asset with a predetermined fixed life span.

The integrated view of goodwill’s contribution embraced by FASB 142 aims to achieve two objectives:

119 H. Kiymaz et al ⎪ The International Journal of Business and Finance Research ♦Vol. 2 ♦ No. 1 ♦ 2008

1) acknowledge goodwill’s economic value in a business combination and 2) design a reporting method that improves the transparency of goodwill’s contribution. The statement mandates an annual test of goodwill impairment in two steps at the reporting unit level. If the unit’s fair value is greater than its carrying amount including goodwill, then goodwill is not impaired and no change in goodwill reporting is required. If the unit’s fair value is less than its carrying amount including goodwill, then the company must recognize that difference on the balance sheet and income statement. Academic debate on goodwill accounting centers on how analysts and investors weigh goodwill in their decision to hold or acquire post- merger company stock.

Applying an event methodology, we conduct an early direct test of that expectation. We use abnormal stock returns on 188 companies that either passed or failed the new annual two-step goodwill impairment test after December 15, 2001, the statement’s effective date.

LITERATURE REVIEW

Prior to the mid-1990s the research literature on post-acquisition goodwill treatment relied on indirect tests of market information efficiency to see whether analysts and investors peered through a firm’s particular accounting treatment to reveal earnings net of goodwill. Conventional wisdom suggested that accounting method alone should not materially affect a post-acquisition firm’s market value (Davis 1990 and 1996, McCarthy 1995). Empirical tests conducted before December 2001 to assess the claim of market transparency supported the notion that analysts, at least, and perhaps market investors saw through the accounting fog of goodwill effects on earnings to reveal the basis for firm value.

More recent indirect studies used alternate measures to test market information efficiency regarding goodwill accounting. Moehrle’s study (2001) echoed another aspect of conventional thinking by showing statistically that cash flow explains company returns as well as traditional accounting earnings methods, net of extraordinary items. Henning’s (2000) investigation optimistically showed that investors appear to distinguish between “core” goodwill that accompanies a merger and “residual” goodwill that investors appear to quickly discount.

Academic debate about goodwill accounting’s affect on transparency for analysts and investors continued as the FASB considered alternate means of post-acquisition disclosure. Recent studies have applied direct methodologies to help clarify earlier mixed goodwill reporting findings. Herz (2001) abandoned traditional firm valuation models and stated that the most direct approach on which to base goodwill impairment testing is the firm’s current stock price. Hopkins (2000) empirically showed that goodwill accounting method and the number of years since the merger did affect analysts’ estimates. Hopkins and his co-authors worried that eliminating the pooling method, thus increasing goodwill expense and lowering net earnings, would depress the affected firms’ stock prices.

Norris (2000) finds that reporting goodwill changes negatively affects market value after the quarterly earnings announcement. Their findings hinted that investors looked more at earnings than at cash flow. Jennings’ (2001) cross-section investigation further suggests that earnings before goodwill amortization is more useful as a summary indicator of share value. Hirschey (2002) bluntly concludes that if goodwill is impaired, the market will recognize it. The movement in the empirical literature from the relatively benign posture that goodwill accounting has little, if any, effect on investor actions and market value seems to be yielding toward an information-based view that goodwill, and especially unrealized expectations regarding goodwill, matters.

Hirschey and Richardson (2003) review the accounting profession’s logical migration on goodwill accounting from APB 16, FASB 141 and FASB 142 to presage future goodwill announcement events and market reaction. They draw their data from 80 goodwill write-off announcements across 32 industries

120 The International Journal of Business and Finance Research ♦ Volume 2 ♦ Number 1 ♦ 2008

occurring from 1992-1996. They applied three separate assessment measures to abnormal returns over “long windows” of 255 days encompassing event dates. Their results indicate the market does react negatively to write-off announcements and that the negative reaction continues for some time after the announcement. However, theirs is a pre-FASB 142 effective date investigation.

The next section describes our direct market valuation test of FASB 142 for a sample of 188 firms across ten industries post-December 15, 2001, separated into two subsets. One subset of firms in the working hypothesis amortized goodwill but passed the impairment test. The other subset of firms in the working hypothesis includes firms that failed the impairment test and wrote off goodwill, in whole or in part, to report lower than expected earnings. The results section shows statistically significant market reaction for certain event windows around the firms’ goodwill announcement date. We interpret the results as information-based market responses to expectation changes from goodwill announcements.

DATA AND METHODOLOGY

We applied two screening criteria to an initial sample of 209 firms reported in Lexis-Nexis making goodwill announcements post-June 30, 2001. First, firms in the final sample should be traded on the New York Stock Exchange (NYSE), American Stock Exchange-National Association of Securities Dealers Automated Quotations (AMEX-NASDAQ), to ensure that stock returns of firms can be

Table 1: Sample Selection Panel A: Sample Frequency by Industry Classification

Characteristic All firms Impairment passed Impairment failed Lexis-Nexis reports 209 106 103 Less: No data on CRSP 13 6 7

Less: Missing data 8 5 3 Net Sample 188 95 93 The panel above shows sample selection frequency by industry classification.

Panel B: Frequency of Sample Firms by Industry Classification

Industry classification N % SIC20-21 Food 11 5.8 SIC26-27 Paper Products/Publishing 10 5.3 SIC28 Chemicals 12 6.5 SIC31-35 Manufacturing 20 10.6 SIC36-38 Electronics/Equipments 36 19.2 SIC48-49 Communications/Utilities 28 14.9 SIC50-58 Retail/Trade 15 7.90 SIC60-63 Financials 10 5.3 SIC73-79 Services 35 18.7

SIC80-87 11 5.8 Total 188 100 % The panel above shows sample selection frequency by industry classification. retrieved from the Center for Research in Security Prices (CRSP) daily return database. Second, firms with missing data on the CRSP database were eliminated. The net sample consisted of 188 firms making goodwill announcements during the period October 22, 2001 to November 27, 2002, shown in panels A and B of Table 1.

121 H. Kiymaz et al ⎪ The International Journal of Business and Finance Research ♦Vol. 2 ♦ No. 1 ♦ 2008

Data in Panel B of Table 1 report the frequency of sample firms by industry classification. Of the total 188 goodwill-related announcements, 36 (19.2%) are in Standard Industrial Classification (SIC) 36-38, 35 (18.7%) are in SIC 70-79, 28 (14.9%) are in SIC 48-49, and 20 (10.6%) are in SIC 31-35. Nearly 45 percent of the firms fell into SIC 31-35 manufacturing, 36-38 electronics, electrical equipment and SIC 48-49 communications and utilities.

The event study methodology measures the abnormal returns--actual company stock return less the regression estimated average market return--for goodwill announcements by the acquiring firms. The single-market model used in the parameter estimation appears as (1) below:

R ti, = α i + β Di, . R tD, + ε ti, (1) Where: Ri,t = the rate of return on security i on day t, αi = the intercept term, βi,D = the slope of the regression line of the firm i's returns against the returns on the market value weighted CRSP Index, RD,t = the rate of return on the market value weighted CRSP Index, εi,t = the residuals.

An abnormal return for common stock of firm i on day t is defined as:

AR = R ti,ti, - Rˆ ti, (2) where,

ˆ Rˆ ti, = αˆi + β . R tD, Di, (3) and αi, and βi,D, are estimated market model parameters obtained by using the pre-estimation period: t = - 316 days to t = - 61 days. We used 255 days to estimate model parameters for the event window to analyze abnormal returns.

We derive cumulative abnormal returns of firm i (CARi) by accumulating ARi,t's over a k-trading period running from day d1 to day d2: d 2 CARi = ∑ AR ti, (4) =t d 1 Average daily abnormal returns (AARs) is obtained by dividing (4) by N: 1 N AARt (= ∑ AR). , ti (5) N =1i

Finally, the Cumulative Average Abnormal Returns (CARs) for a sample of N firms across a k-day event window we calculated as follows:

1 d 2 N (=CAR ) ∑∑ AR ti, (6) N =t d 1 =1i

The expected values of abnormal returns and average abnormal returns are zero in the absence of abnormal performance. The test for significance follows Brown and Warner (1985). We divided the net total sample along FASB 142 criteria to reveal that 95 firms passed and 93 firms failed the two-step goodwill impairment test criteria.

122 The International Journal of Business and Finance Research ♦ Volume 2 ♦ Number 1 ♦ 2008

Table 2: Abnormal Returns Surrounding Goodwill Announcements Panel A: Average Daily Abnormal Returns (AARs)

Firms: N=188 Days AARs (%) t-value Positive: Negative Generalized Sign Test -10 -0.02 -2.57*** 83:105 -1.22 -9 -0.56 -1.34 83:105 -1.22 -8 -0.21 -0.8 91:97 -0.05 -7 -0.06 -0.45 96:92 0.68 -6 -0.2 -1.2 91:97 -0.05 -5 0.41 -0.16 81:107 -1.51 -4 0.09 -0.65 80:108 -1.65 -3 -0.26 -1.45 92:96 0.1 -2 -0.2 -1.22 93:95 0.24 -1 0.04 -0.33 99:89 1.12 0 -0.51 -1.93* 81:107 -1.51 1 -0.67 -0.96 88:100 -0.49 2 -0.71 -1.73* 91:97 -0.05 3 -0.33 -0.45 100:88 1.27 4 -0.28 -0.9 94:94 0.39 5 -0.04 0.42 103:85 1.7 6 0.19 -0.72 93:95 0.24 7 0.34 0.46 95:93 0.54 8 0.27 -0.02 93:95 0.24 9 -0.43 -0.76 84:104 -1.07 10 0.57 0.48 91:95 0.09 The panel above shows average daily abnormal returns (AARs). Panel B: Cumulative Abnormal Returns (CARs)

Firms: N=188 Windows CARs (%) t-value Positive: Negative Generalized Sign Test (-1, 0) -0.47 -1.6 84:104 -1.07 (-1, 1) -1.13 -1.86* 84:104 -1.07 - (-5, 5) -2.46 2.83*** 89:99 -0.34 - (-10, 10) -2.57 3.56*** 80:108 -1.65 The panel above shows cumulative abnormal returns (CARs). ***, **, and * note significance at the 1%, 5% and 10% levels, respectively. This table presents the abnormal return to firms surrounding the announcement of goodwill write-offs. The null hypothesis is that the average abnormal returns are not statistically different from zero.

EMPIRICAL RESULTS

The wealth gains of goodwill announcements are calculated by using a market model. The results for the aggregate sample are reported in Table 2. The behavior of the daily abnormal returns to firms with goodwill impairment tests during the 21-day period surrounding the announcement based on a market model is reported in Panel A of Table 2. The average abnormal returns (AARs) are 0.04% and -0.51% on the days -1 and 0 and only results for day 0 are statistically significant at 10%. Furthermore, on the announcement day a majority of firms experience negative abnormal returns (107 out of 188 firms). The cumulative abnormal returns (CARs) may provide a better picture of stock market reactions and hence we report results for various event windows shown in Panel B of Table 2. For the (-1,0) and (-1,+1) windows,

123 H. Kiymaz et al ⎪ The International Journal of Business and Finance Research ♦Vol. 2 ♦ No. 1 ♦ 2008

the CARs are -0.47% and -1.13% respectively, but only the CARs on the latter window are weakly significant. When we analyze longer event windows, the findings are strongly negative. For example, the CARs for the window (-10, +10) is -2.57% and are statistically significant at the 1% level.

We, then, divide the sample based on the impairment test results. Data in Panel A of Table 3 below show AARs for each firm subset—those passing and those failing the FASB 142 goodwill impairment test.

Table 3: Abnormal Returns Based on Impairment Test Panel A: Average Daily Abnormal Returns and t-test Values

Impairment passed: N=95 Impairment failed: n=93 Days AARs (%) t-value AAR (%) t-value -10 0.42 -1.05 -0.47 -2.59*** -9 -0.49 -0.69 -0.63 -1.21 -8 -0.12 -0.35 -0.30 -0.78 -7 0.14 0.19 -0.27 -0.45 -6 -0.66 -1.50 0.27 -0.19 -5 1.01 0.53 -0.21 -0.77 -4 -0.15 -0.44 0.33 -0.48 -3 -0.03 -0.48 -0.50 -1.58 -2 -0.06 -0.44 -0.35 -1.29 -1 0.16 -0.06 -0.08 -0.41 0 -0.50 -0.81 -0.52 -1.94* +1 -1.38 -2.03** 0.06 0.69 +2 -1.27 -2.07** -0.13 -0.37 +3 -0.93 -0.60 0.29 -0.04 +4 -0.15 -0.70 -0.41 -0.58 +5 -0.03 0.40 -0.04 0.19 +6 0.87 1.49 -0.50 -2.53** +7 0.55 0.64 0.12 0.00 +8 1.25 1.07 -0.73 -1.11 +9 -1.33 -1.88* 0.50 0.82 +10 1.24 1.44 -0.09 -0.75 The above panel shows average daily abnormal returns and t-test values. ***, **, and * note significance at the 1%, 5% and 10% levels. Panel A: Cumulative Abnormal Returns and t-test Values

Impairment passed: N=95 Impairment failed: n=93 Windows CARs (%) t-value CARs (%) t-value (-1, 0) -0.33 -0.61 -0.60 -1.66 (-1, 1) -1.72 -1.67 -0.54 -0.96 (-5, 5) -3.33 -2.02** -1.57 -1.98* (-10, 10) -1.48 -1.69* -3.69 -3.35*** The above panel shows cumulative abnormal returns and t-test values. ***, **, and * note significance at the 1%, 5% and 10% levels. The table shows the abnormal return to firms surrounding the goodwill announcement date. The sample is divided in two sets based on the impairment test results. The null hypothesis is that average abnormal returns are not statistically different from zero.

Both subsets experienced negative stock market reaction on the announcement day, and some firms that either passed the impairment test or failed the impairment test show statistically significant returns. For the subset of firms passing the impairment test, significant results appeared for Days +1 and +2, post- goodwill announcement event date. Absent firm-specific information, we interpret the results to mean

124 The International Journal of Business and Finance Research ♦ Volume 2 ♦ Number 1 ♦ 2008

investors reacted positively to the news. For the subset of firms failing the impairment test, significant AARs are reported for Day –10 and Day +6. Weakly significant AARs for the event date also appear for these firms. We interpret these findings to mean investors either anticipated negative goodwill news early and/or reacted to the negative goodwill news once they received the information.

Data in the Panel B of Table 2 report Cumulative Abnormal Returns (CARs) for each firm subset—those passing and those failing the FASB 142 goodwill test. The CARs measure captures more of the market reaction and serves as a more inclusive measure of investor reaction. For the 95 firms passing the impairment test no CAR window set of returns is significant above 10 percent, a statistically weak criterion. We interpret this finding to suggest that investors’ expectations were confirmed. Since annual goodwill write-downs were not altered, the reporting unit passed the impairment test inducing little or no market reaction based on that information.

Table 4: The Impact of Exchange Traded and Industry Affiliation on Abnormal Returns Panel A: Abnormal Returns by Exchange Traded Industry

NYSE: n=86 AMEX+NASDAQ: n=102 Days AARs (%) t-value AAR (%) t-value -10 -1.13 -4.06*** 0.91 0.24 -1 -0.58 -1.62 0.57 1.04 0 -0.05 -0.02 -0.89 -2.61** +1 -0.44 0.15 -0.86 -1.44 +10 0.21 0.17 0.89 0.50

Windows CARs (%) t-value CARs (%) t-value (-1, 0) -0.64 -1.16 -0.32 -1.11 (-1, 1) -1.08 -0.86 -1.18 -1.74* (-5, 5) -2.39 -1.98* -2.51 -2.02** (-10, 10) -3.76 -3.15*** -1.57 -1.94* The above panel shows abnormal returns by exchange traded industry. ***, **, and * note significance at the 1%, 5% and 10% levels respectively. Panel B: Abnormal Returns by Industry Affiliation

Windows (-1, 0) (-1, 1) (-5, 5) (-10, 10) CAR (%) t-value CAR (%) t-value CAR (%) t-value CAR (%) t-value SIC 20-21 2.19 -0.71 2.35 -0.61 4.47 -1.23 4.55 -0.77 SIC 25-27 -1.26 -1.13 -2.79 -1.94* -5.43 1.99* -7.86 -2.08** SIC 28 2.79 3.59*** 2.79 3.60*** -5.27 -0.28 -11.12 -1.24 SIC 31-35 -1.17 -1.30 -4.50 -2.54** -6.97 -2.79** -8.76 -2.07** SIC 36-38 0.32 -0.42 -0.60 -0.77 -3.85 2.32** -7.71 -3.79*** SIC 48-49 -3.74 -2.89** -6.15 -3.20*** -10.44 -2.77** -3.55 -1.64 SIC 50-59 -0.93 -0.32 0.30 1.23 7.98 2.84** 9.41 2.11** SIC 60-63 0.20 -0.19 -0.36 -0.53 -6.14 -2.12* -10.44 -2.49** SIC 70-79 -1.64 -1.94* -1.11 -1.49 1.58 -0.40 2.68 -0.22 SIC 80-87 4.84 -1.12 6.99 -1.43 5.74 -0.48 9.11 -0.36 The above panel shows abnormal returns by industry affiliation. ***, **, and * note significance at the 1%, 5% and 10% levels respectively. In contrast, firms failing the goodwill impairment test, show highly significant and negative CARs for the +/- 10 day event window. Investors apparently respond to announcements revealing goodwill write-off due to the FASB 142 test. In general, results from our direct test provide evidence that announced changes in goodwill write-offs do inform investors, some of whom react by selling company stock, likely in the face of unexpected goodwill impairment. For companies passing the FASB 142 goodwill test, the lack of new

125 H. Kiymaz et al ⎪ The International Journal of Business and Finance Research ♦Vol. 2 ♦ No. 1 ♦ 2008

information flowing to investors prompts no or muted market response. We emphasize that the CARs data results for each firm group, those passing as well as those failing the FASB 142 criteria, are logically consistent with a market information efficiency view.

In Table 4, we report stock reaction of firms based on both the exchange in which stock is traded and industry affiliation of firms. Panel A of Table 4 reports the variation in stock price reaction based on the exchange listing. The majority of firms (102 out of 188) are traded in AMEX+NASDAQ. Since firms traded on NYSE are larger, we also attempt to see whether the stock price reaction would differ with respect to sizes of firms. On the announcement date, the AARs for AMEX+NASDAQ listed group is -0.89% and statistically significant at 1% level. The CARs, on the other hand, are negative and statistically significant for both groups of firms. In Panel B of Table 4, we analyze the industry affiliation of firms and abnormal returns based on the primary SIC of firms. The reaction seems to vary among industries. While some industries show positive reactions, others show negative reactions. For example, during the event window (-1,0), the abnormal returns are positive and statistically significant in SIC 28 (2.79%) and SIC80-87 (4.84%), while they are negative in most of the SIC groups. The SIC 48-49 experience the highest negative returns of -3.74%, followed by - 1.64% by SIC70-79 group.

Our results mirror those reported by Hirschey and Richardson (2003) who applied a similar methodology to their 1992-1996 cross-section of prior merged firms. Our sample of 188 draws from 25 of the 32 industries used in the Hirschey sample of 80 firms. They report statistically significant CARs on event day announcements for companies reporting negative goodwill write-offs where our findings reflect a slightly larger window to reveal significantly negative CARs for firms with goodwill write-off announcements. Further, they report weakly significant event day results for write-offs from companies reporting positive earnings while we report no significant event day CARs. Differences between these sets of findings could be due to the market’s general mood and the speed of investor response, given the goodwill information; optimistic during Hirschey’s 1992-1996 data collection period and pessimistic during our 2001-2002 data collection period.

SUMMARY AND CONCLUSION

Our investigation is an early direct test of the FASB 142 effect on market response to post-merger goodwill announcements. Applying a standard event methodology, we derived daily and cumulative abnormal returns for 188 firms making goodwill announcements between October 22, 2001 and November 27, 2002. We divided the firms into those that passed versus failed the two-part FASB 142 goodwill impairment test and derived both daily and cumulative abnormal return measures for each subset. Cumulative abnormal return results show that investors react strongly to negative goodwill announcements over a –10 to +10 day window around the event date. For firms passing the impairment test, cumulative abnormal return results are only weakly significant for the same –10/+10 day window. The findings further show that the abnormal returns are negative for NYSE and AMEX+NASDAQ listed stocks. When we analyze the industry affiliation of firms and abnormal returns based on the primary SIC of firms, we find that the market reactions to goodwill announcements vary among industries. While some industries show positive reactions, others show negative reactions.

126 The International Journal of Business and Finance Research ♦ Volume 2 ♦ Number 1 ♦ 2008

REFERENCES

Brown, S.J. & Warner, J.B. (1985) “Using daily stock returns: The case of event studies,” Journal of Financial Economics, vol. 14, p. 3-31.

Davis, Michael (1990) “Differential Market Reaction to Pooling and Purchase Methods,” The Accounting Review, July, p. 696-709. Davis, Michael (1996) “The Purchase vs. Pooling Controversy: How the Stock Market Responds to Goodwill,” Journal of Applied Corporate Finance, Spring, p. 50-59.

Henning, Steven L., Barry Lewis and Wayne Shaw (2000) “Research Reports—Valuation of the Components of Purchased Goodwill,” Journal of Accounting Research, vol. 38, Autumn, p. 375-387.

Herz, Robert H., Teresa E. Iannaconi, Laureen A. Maines, Krishna Palepu, Stephen G. Ryan, Katherine Schipper, Catherine M. Schrand, Douglas J. Skinner and Linda Vincent (2001) “Equality Valuation Models and Measuring Goodwill Impairment,” Accounting Horizons, vol. 15, June, p. 161-171.

Hirschey, Mark and Vernon Richardson (2002) “Information Content of Accounting Goodwill Numbers,” Journal of Accounting and Public Policy, Fall, p. 173-193.

Hirschey, Mark and Vernon Richardson (2003) “Investor Underreaction to Goodwill Write-Offs,” Financial Analysts Journal, November-December, p. 75-84.

Hopkins, Patrick, Richard Houston and Michael Peters (2000) “Purchase, Pooling, and Equity Analysts’ Valuation Judgments,” The Accounting Review, vol. 75. July, p. 257-281.

Jennings, Ross, Marc LeClere and Robert Thompson (2001) “Goodwill Amortization and the Usefulness of Earnings,” The Financial Analysts Journal, September-October, p. 20-29.

McCarthy, Mark G. and Douglas Schneider (1995) “Market Perception of Goodwill: Some Empirical Evidence,” Accounting and Business Research, vol. 26, Winter, p. 69-82.

Moehrle, Stephen and James Wallace (2000) “Everyone Out of the Pool,” Journal of Accountancy, May, p. 189-196.

Moehrle, Stephen and Jennifer Reynolds-Moehrle (2001) “Say Good-bye to Pooling and Goodwill Amortization,” Journal of Accountancy, September, p. 31-38.

Moehrle, Stephen R, Jennifer Reynolds-Moehrie and James Wallace (2001) “How Informative Are Earnings Numbers that Exclude Goodwill Amortization?” Accounting Horizons, September, p. 243-259.

Norris, Kathleen B. and Frances Ayres (2000) “Security Market Reaction to Purchase Business Combinations at the First Earnings Statement,” Journal of Business Research, July, p. 79-88. The Wall Street Journal Corporate Index

127